Enhancing the thermal stability of polythiophene:fullerene solar cells by decreasing effective polymer regioregularity

Article abstract of DOI:10.1021/ja064434r

Regioregularity has been shown to be an important aspect for the properties of poly(3-hexylthiophene) both in the pristine material and in bulk heterojunction photovoltaics with the fullerene derivative PCBM. Here we present a straightforward method to adjust the effective regioregularity of poly(3-hexylthiophene) by introducing 3,4-dihexylthiophene into the polymer chain. By investigating bulk heterojunction morphology and photovoltaic performance upon annealing, we observe that polythiophene with 91% regioregularity maintains equivalent electronic properties but forms a more thermally stable interpenetrating network with PCBM than polythiophene with regioregularity >96%. Copyright

Full text of DOI:10.1021/ja064434r

Published on Web 10/10/2006
Enhancing the Thermal Stability of Polythiophene:Fullerene Solar Cells by
Decreasing Effective Polymer Regioregularity
Kevin Sivula, Christine K. Luscombe, Barry C. Thompson, and Jean M. J. Fre´chet*
Departments of Chemistry and Chemical Engineering, UniVersity of California, Berkeley, 718 Latimer Hall,
Berkeley, California 94720-1460, and Materials Sciences DiVision, Lawrence Berkeley National Laboratory,
Berkeley, California 94720
Received June 22, 2006; E-mail: frechet@berkeley.edu
Scheme 1. Synthetic Route for Lowering Regioregularity by the
Incorporation of a Dihexylthiophene Repeat Unit
Thin-film photovoltaics made from a bulk heterojunction (BHJ)
between poly(3-hexylthiophene) (P3HT) and the fullerene derivative
[6,6]-phenyl C61 butyric acid methyl ester (PCBM) have emerged
as the most promising organic solar cell system.1,2 The morphology
of the active layer in these devices has garnered particular interest
because electron transfer from the donating polymer to the accepting
fullerene3 occurs over a distance of ca. 10 nm, suggesting that an
interpenetrating network of that length scale is necessary for
maximum device performance. A thermal annealing step, above
the Tg of the polymer, is necessary to increase its crystallinity in
order to improve its electrical properties, but thermal annealing also
drives the phase segregation of P3HT and PCBM.4,5 Phase
segregation destroys the BHJ, thereby reducing device perfor-
mance,5 and thus, optimal thermal annealing conditions are required
to delicately balance increasing P3HT crystallinity while minimizing
phase segregation of the active components. The recent discovery
of conditions to attain a thermally stable interpenetrating network
of P3HT and PCBM1 has led to devices with simulated solar power
(AM 1.5G) conversion efficiencies in the range of 4-5% using a
90-93% regioregular (RR) P3HT from a commercial source.6
Much effort has been devoted to understanding how the
properties of P3HT affect its electronic characteristics. Increasing
the molecular weight (MW) of the polymer has been shown to
enhance the charge carrier mobility of pristine P3HT in both the
field effect7 and space-charge limited8 regimes. Regioregularity has
also been shown to affect charge carrier mobility in polythiophene
thin films.9 As expected, P3HT with both increased molecular
weight10 and regioregularity11 enhances performance in P3HT:
PCBM photovoltaic devices. While tuning the MW of P3HT is
easily achieved through modulating the polymerization conditions,
and methods of producing essentially 100% RR-P3HT have been
reported,12,13 a straightforward route to adjusting the regioregularity
of P3HT has been elusive. We now present a new method to tune
the regioregularity of P3HT and report that, surprisingly, a small
decrease in the effective regioregularity of P3HT appears to confer
more thermal stability to the BHJ of P3HT and PCBM with
implications on long-term performance.
respectively). MALDI-TOF MS analysis of the purified copolymer
1
confirms the incorporation of monomer 1, and H NMR charac-
terization (Figures S1 and S2) allows the comparison of regioregu-
larities. Regioregularity is typically calculated by comparing the
integrated peaks corresponding to the R-methylene protons on the
hexyl chains in head-to-tail (HT) versus head-to-head (HH)
linkages,12 δ2.78 and δ2.54, respectively. This gives >96% HT
for the control P3HT. The same analysis gives an effective
regioregularity of 91% for poly(1-co-2), which is lower, as expected,
since the incorporation of monomer 1 increases the content of non-
HT linkages in the copolymer.
To investigate the BHJ morphology, thin films of poly(1-co-2),
or the control P3HT, and PCBM (1:1 by weight) were spin-cast
from chlorobenzene onto a freshly cleaved single-crystal NaCl
substrate and annealed, with the top unconfined, at typical device
fabrication conditions (150 °C for 30 min). The films were then
removed by floating on water and placed on TEM grids for analysis.
The resulting micrographs (Figure 1) show the typical large-scale
phase segregation of the PCBM from the P3HT in the BHJ prepared
with the control polymer, along with no obvious features upon close
inspection of the regions free from PCBM (Figure 1a). In contrast,
blends made with poly(1-co-2) produced a thermally stable
interpenetrating network with vermicular features on the order of
20 nm in width (Figure 1b). This morphology was also observed
in BHJs prepared using the commercially available polymer.1 We
suggest that the high driving force for crystallization of the RR-
P3HT control material initiates the phase segregation of the BHJ
by the exclusion of PCBM from the ordered P3HT domains. The
stark contrast in the ability of poly(1-co-2) to retain a bicontinuous
morphology relative to RR-P3HT is consistent with our claim that
introducing disorder, in the form of the dihexyl-functionalized
monomer 1, serves to attenuate the driving force for crystallization
and facilitates the retention of a thermally stable BHJ during
annealing. Unoptimized photovoltaic devices fabricated with the
control polymer or poly(1-co-2) and PCBM (1:1 by weight) support
this conjecture (Figure 2; note that each data point is the average
for eight devices). While 30 min of annealing after cathode
deposition produced devices with similar power conversion ef-
ficiencies (up to ηPC ) 4.3 and 4.4% for the control and poly(1-
To reduce the effective regioregularity of P3HT, we began with
the McCullough method, which is known to produce highly RR-
P3HT,14 while also randomly incorporating a thiophene monomer
with n-hexyl chains at both the 3 and 4 positions. Monomers 1
and 2 (Scheme 1) were prepared using standard procedures (see
Supporting Information). The activated zinc species 1* and 2* were
prepared in separate flasks at known concentrations before 1* was
combined with 2* in situ in the ratio of 1:24 prior to the addition
of 1 mol % of nickel catalyst to initiate polymerization. A control
sample of RR-P3HT was synthesized using the same procedure,
and the molecular weights of the polymers were found to be similar
(GPC Mn ) 22 and 28 kDa for poly(1-co-2) and the control P3HT,
9
13988
J. AM. CHEM. SOC. 2006, 128, 13988-13989
10.1021/ja064434r CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
co-2), we can conclude that incorporation of a small amount of
monomer 1 into P3HT does not significantly effect its electronic
properties. Additionally, since we have observed identical device
performance and stability for poly(1-co-2) and the commercially
available material, which has a significantly higher Mn,6 it is unlikely
that molecular weight plays a major role in forming a stable BHJ.
However, because poly(1-co-2) and the commercial material have
similar amounts of disorder (in the form of HH linkages), greater
than the control RR-P3HT, we conclude this is an important aspect
in forming a thermally stable BHJ with PCBM. Thus the 5%
efficiency1 obtained through rigorous optimization of devices
prepared using commercial P3HT may be due to the fortuitous lower
RR of that material,6 which eliminates the concern of phase
segregation. Obtaining a stable interpenetrating network of electron
donor and electron acceptor is important for organic photovoltaics.
Here we have seen that, while maximizing regioregularity in P3HT
may provide the best electronic properties in the pristine material,
the most stable photovoltaic performance is obtained with a
polythiophene with a tuned regioregularity to control the BHJ
morphology.
Acknowledgment. Financial support of this work by the
Department of Energy (DE-AC03-76SF00098) is gratefully ac-
knowledged. B.C.T. acknowledges funding from an ACS PRF
postdoctoral AEF. We thank Prof. A. P. Alivisatos and his group
for fabrication facilities and helpful discussion of this work.
Figure 1. Transmission electron micrographs of annealed BHJ thin films
containing 1:1 by weight RR-P3HT:PCBM (a) and poly(1-co-2):PCBM
(b). An interpenetrating network of polymer and fullerene is observed with
poly(1-co-2) (b), while no obvious features are seen in the polymer region
of the control case (a). The insets show a wider view of the films and the
micron-scale phase segregation of PCBM (dark region) from the RR-P3HT
in the control case. No micron-scale features are observed when poly(1-
co-2) is used in the BHJ. The film edge is shown to confirm focus in the
inset of (b).
Supporting Information Available: Complete synthetic proce-
dures, characterization, and device fabrication methods. This material
is available free of charge via the Internet at http://pubs.acs.org.
References
(1) Ma, W. L.; Yang, C. Y.; Gong, X.; Lee, K.; Heeger, A. J. AdV. Funct.
Mater. 2005, 15, 1617-1622.
(2) (a) Li, G.; Shrotriya, V.; Huang, J. S.; Yao, Y.; Moriarty, T.; Emery, K.;
Yang, Y. Nat. Mater. 2005, 4, 864-868. (b) Reyes-Reyes, M.; Kim, K.;
Carroll, D. L. Appl. Phys. Lett. 2005, 87, 083506.
(3) (a) Markov, D. E.; Amsterdam, E.; Blom, P. W. M.; Sieval, A. B.;
Hummelen, J. C. J. Phys. Chem. A 2005, 109, 5266-5274. (b) Theander,
M.; Yartsev, A.; Zigmantas, D.; Sundstrom, V.; Mammo, W.; Andersson,
M. R.; Inganas, O. Phys. ReV. B: Condens. Matter 2000, 61, 12957-
12963.
(4) (a) Chirvase, D.; Parisi, J.; Hummelen, J. C.; Dyakonov, V. Nanotech-
nology 2004, 15, 1317-1323. (b) Savenije, T. J.; Kroeze, J. E.; Yang, X.
N.; Loos, J. AdV. Funct. Mater. 2005, 15, 1260-1266.
(5) Sivula, K.; Ball, Z. T.; Watanabe, N.; Fre´chet, J. M. J. AdV. Mater. 2006,
18, 206-210.
(6) Poly(3-hexylthiophene-2,5-diyl) from Rieke Metals, Inc. Our analysis
confirms the advertised RR ) 90-93% DCB GCP: Mn ) 31 kDa, PDI
) 2.0.
(7) (a) Zen, A.; Pflaum, J.; Hirschmann, S.; Zhuang, W.; Jaiser, F.;
Asawapirom, U.; Rabe, J. P.; Scherf, U.; Neher, D. AdV. Funct. Mater.
2004, 14, 757-764. (b) Kline, R. J.; McGehee, M. D.; Kadnikova, E. N.;
Liu, J. S.; Fre´chet, J. M. J.; Toney, M. F. Macromolecules 2005, 38, 3312-
3319.
Figure 2. Performance of photovoltaic devices fabricated with the control
RR-P3HT or poly(1-co-2). The eight-device average power conversion
efficiency (ηPC) at standard conditions as a function of annealing time at
150 °C and typical J-V behavior (insets) observed after 30 and 300 min
of annealing illustrate the difference between the materials. The error bars
represent the 95% confidence interval of the average values.
(8) Goh, C.; Kline, R. J.; McGehee, M. D.; Kadnikova, E. N.; Fre´chet, J. M.
J. Appl. Phys. Lett. 2005, 86, 122110.
(9) (a) Sirringhaus, H.; Brown, P. J.; Friend, R. H.; Nielsen, M. M.; Bechgaard,
K.; Langeveld-Voss, B. M. W.; Spiering, A. J. H.; Janssen, R. A. J.; Meijer,
E. W.; Herwig, P.; de Leeuw, D. M. Nature 1999, 401, 685-688. (b)
Pandey, S. S.; Takashima, W.; Nagamatsu, S.; Endo, T.; Rikukawa, M.;
Kaneto, K. Jpn. J. Appl. Phys., Part 2 2000, 39, L94-L97.
co-2), respectively, at AM 1.5G, 100 mW cm-2), continued
annealing of RR-P3HT devices causes a decrease in performance,
and after 300 min at 150 °C, the average ηPC fell to 2.6%. Subjecting
devices with a poly(1-co-2):PCBM active layer to the same
extended annealing conditions does not strongly effect the average
device performance, and the eight-device average ηPC remained at
3.5%. We observe comparable annealing behavior with the com-
mercial P3HT (see Supporting Information). In addition, we note
the decrease in device performance in the RR-P3HT devices,
caused by phase segregation, occurs at longer times than for the
TEM samples due to the confining effects of the Al cathode.5
Given that we have been able to obtain similar performance at
short annealing times with both the control polymer and poly(1-
(10) Schilinsky, P.; Asawapirom, U.; Scherf, U.; Biele, M.; Brabec, C. J. Chem.
Mater. 2005, 17, 2175-2180.
(11) Kim, Y.; Cook, S.; Tuladhar, S. M.; Choulis, S. A.; Nelson, J.; Durrant,
J. R.; Bradley, D. D. C.; Giles, M.; McCulloch, I.; Ha, C. S.; Ree, M.
Nat. Mater. 2006, 5, 197-203.
(12) Chen, T.-A.; Wu, X.; Rieke, R. D. J. Am. Chem. Soc. 1995, 117, 233-
244.
(13) (a) Loewe, R. S.; Khersonsky, S. M.; McCullough, R. D. AdV. Mater.
1999, 11, 250-253. (b) Miyakoshi, R.; Yokoyama, A.; Yokozawa, T. J.
Am. Chem. Soc. 2005, 127, 17542-17547.
(14) Liu, J.; McCullough, R. D. Macromolecules 2002, 35, 9882-9889.
JA064434R
9
J. AM. CHEM. SOC. VOL. 128, NO. 43, 2006 13989